CN117498262A - High-voltage direct-current electronic load switch protection circuit - Google Patents
High-voltage direct-current electronic load switch protection circuit Download PDFInfo
- Publication number
- CN117498262A CN117498262A CN202311422400.7A CN202311422400A CN117498262A CN 117498262 A CN117498262 A CN 117498262A CN 202311422400 A CN202311422400 A CN 202311422400A CN 117498262 A CN117498262 A CN 117498262A
- Authority
- CN
- China
- Prior art keywords
- time sequence
- electronic load
- current
- sequence
- vector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 239000013598 vector Substances 0.000 claims abstract description 78
- 238000006243 chemical reaction Methods 0.000 claims abstract description 28
- 238000010586 diagram Methods 0.000 claims description 54
- 238000012800 visualization Methods 0.000 claims description 31
- 239000011159 matrix material Substances 0.000 claims description 25
- 238000000605 extraction Methods 0.000 claims description 20
- 238000012545 processing Methods 0.000 claims description 15
- 230000011218 segmentation Effects 0.000 claims description 14
- 238000005457 optimization Methods 0.000 claims description 12
- 238000013527 convolutional neural network Methods 0.000 claims description 9
- 230000004913 activation Effects 0.000 claims description 4
- 238000003062 neural network model Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 238000011176 pooling Methods 0.000 claims description 2
- 238000012546 transfer Methods 0.000 claims description 2
- 230000002159 abnormal effect Effects 0.000 abstract description 11
- 238000013135 deep learning Methods 0.000 abstract description 11
- 238000005516 engineering process Methods 0.000 abstract description 8
- 238000013473 artificial intelligence Methods 0.000 abstract description 6
- 238000000034 method Methods 0.000 description 31
- 230000008901 benefit Effects 0.000 description 11
- 230000008569 process Effects 0.000 description 11
- 238000004422 calculation algorithm Methods 0.000 description 9
- 238000004458 analytical method Methods 0.000 description 6
- 230000009471 action Effects 0.000 description 5
- 238000012360 testing method Methods 0.000 description 5
- 230000002596 correlated effect Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000012512 characterization method Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 3
- 238000004148 unit process Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 230000001052 transient effect Effects 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000875 corresponding effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000013021 overheating Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H1/00—Details of emergency protective circuit arrangements
- H02H1/0092—Details of emergency protective circuit arrangements concerning the data processing means, e.g. expert systems, neural networks
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/766—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using regression, e.g. by projecting features on hyperplanes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H1/00—Details of emergency protective circuit arrangements
- H02H1/0007—Details of emergency protective circuit arrangements concerning the detecting means
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H3/00—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal electric working condition with or without subsequent reconnection ; integrated protection
- H02H3/08—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal electric working condition with or without subsequent reconnection ; integrated protection responsive to excess current
- H02H3/087—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal electric working condition with or without subsequent reconnection ; integrated protection responsive to excess current for dc applications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H5/00—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal non-electric working conditions with or without subsequent reconnection
- H02H5/04—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal non-electric working conditions with or without subsequent reconnection responsive to abnormal temperature
- H02H5/047—Emergency protective circuit arrangements for automatic disconnection directly responsive to an undesired change from normal non-electric working conditions with or without subsequent reconnection responsive to abnormal temperature using a temperature responsive switch
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Emergency Protection Circuit Devices (AREA)
Abstract
A high voltage direct current electronic load switch protection circuit, which obtains current values of electronic loads at a plurality of preset time points in a preset time period; arranging the current values of the electronic loads at a plurality of preset time points into time sequence input vectors of the current values of the electronic loads according to the time dimension; performing image conversion on the electronic load current value time sequence input vector to obtain a sequence of electronic load current value local time sequence images; extracting current fluctuation time sequence characteristics in the sequence of the electronic load current value local time sequence image; based on the current ripple timing characteristics, it is determined whether to close the switch. Therefore, the real-time current waveform of the electronic load is analyzed by combining with the artificial intelligence technology based on deep learning, and the fluctuation mode characteristics of the electronic load are extracted, so that abnormal conditions are identified, and whether a switch is closed or not is judged, and misoperation caused by outliers is avoided.
Description
Technical Field
The application relates to the technical field of intelligent switch protection circuits, and more particularly, to a high-voltage direct-current electronic load switch protection circuit.
Background
The high-voltage direct-current electronic load can simulate various load conditions and evaluate the stability, efficiency, transient response and the like of the power supply. However, the hvth electronic load generates a large amount of heat during operation, and if heat is not dissipated in time, the temperature of the electronic load increases and even is damaged. In addition, the output current of the high-voltage direct-current power supply may also be fluctuated or suddenly changed, so that the overcurrent of the electronic load is caused, and the service life and the safety of the electronic load are also influenced.
At present, a common high-voltage direct-current electronic load switch protection circuit is to connect a switch in series with the input end of an electronic load, detect the working state of the electronic load through a temperature sensor or a current sensor, and close the switch and cut off the power supply when the temperature or the current exceeds a preset threshold value, thereby protecting the electronic load. This approach, while simple, reliable and quick in response, has some problems. For example, since the threshold is set to be fixed, false or missed conditions may occur, such as outliers due to sudden current peaks or other anomalies that may not represent a true overheat or overcurrent condition. Therefore, an optimized high voltage dc electronic load switch protection circuit is desired.
Disclosure of Invention
The present application has been made in order to solve the above technical problems. The embodiment of the application provides a high-voltage direct-current electronic load switch protection circuit, which acquires current values of electronic loads at a plurality of preset time points in a preset time period; arranging the current values of the electronic loads at a plurality of preset time points into time sequence input vectors of the current values of the electronic loads according to the time dimension; performing image conversion on the electronic load current value time sequence input vector to obtain a sequence of electronic load current value local time sequence images; extracting current fluctuation time sequence characteristics in the sequence of the electronic load current value local time sequence image; based on the current ripple timing characteristics, it is determined whether to close the switch. Therefore, the real-time current waveform of the electronic load is analyzed by combining with the artificial intelligence technology based on deep learning, and the fluctuation mode characteristics of the electronic load are extracted, so that abnormal conditions are identified, and whether a switch is closed or not is judged, and misoperation caused by outliers is avoided.
In a first aspect, a high voltage dc electronic load switch protection circuit is provided, comprising:
a current value acquisition module for acquiring current values of the electronic load at a plurality of preset time points in a preset time period;
the vector arrangement module is used for arranging the current values of the electronic loads at a plurality of preset time points into time sequence input vectors of the current values of the electronic loads according to the time dimension;
the image conversion module is used for carrying out image conversion on the electronic load current value time sequence input vector so as to obtain a sequence of electronic load current value local time sequence images;
the time sequence feature extraction module is used for extracting current fluctuation time sequence features in the sequence of the electronic load current value local time sequence images;
and the switch control module is used for determining whether to close the switch or not based on the current fluctuation time sequence characteristic.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments or the description of the prior art will be briefly introduced below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a block diagram of a high voltage dc electronic load switch protection circuit according to an embodiment of the present application.
Fig. 2 is a flowchart of a method for protecting a hvdc electronic load switch according to an embodiment of the present application.
Fig. 3 is a schematic diagram of a protection method architecture for a hvdc electronic load switch according to an embodiment of the present application.
Fig. 4 is an application scenario diagram of a hvth electronic load switch protection circuit according to an embodiment of the application.
Detailed Description
The following description of the technical solutions in the embodiments of the present application will be made with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
Unless defined otherwise, all technical and scientific terms used in the examples of this application have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the present application is for the purpose of describing particular embodiments only and is not intended to limit the scope of the present application.
In the description of the embodiments of the present application, unless otherwise indicated and defined, the term "connected" should be construed broadly, and for example, may be an electrical connection, may be a communication between two elements, may be a direct connection, or may be an indirect connection via an intermediary, and it will be understood by those skilled in the art that the specific meaning of the term may be understood according to the specific circumstances.
It should be noted that, the term "first\second\third" in the embodiments of the present application is merely to distinguish similar objects, and does not represent a specific order for the objects, it is to be understood that "first\second\third" may interchange a specific order or sequence where allowed. It is to be understood that the "first\second\third" distinguishing objects may be interchanged where appropriate such that the embodiments of the present application described herein may be implemented in sequences other than those illustrated or described herein.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
A hvdc electronic load is a device for simulating various load conditions and evaluating and testing a power supply, and is generally used for testing and verifying the performance and stability of a hvdc power supply, and evaluating the transient response and efficiency of the power supply in an actual operating environment. The main function of the high-voltage direct-current electronic load is to simulate the load condition, and the electric energy output by the power supply is absorbed and converted into heat energy to dissipate heat. It can provide adjustable current and voltage loads to meet different test requirements. By adjusting the current and voltage loads, different operating states and load conditions, such as constant current loads, constant voltage loads, constant power loads, etc., can be simulated.
The application fields of the high-voltage direct-current electronic load are wide, including renewable energy systems such as power electronics, electric automobiles, solar energy and wind energy, battery testing and evaluation, power research and development and testing and the like, and the high-voltage direct-current electronic load can help engineers and researchers evaluate the performance and stability of a power supply, optimize the design of the power supply, improve the efficiency of the system and ensure the reliability and safety of the power supply under various load conditions.
A high voltage dc electronic load switch protection circuit is a circuit for preventing overheating or overcurrent of an electronic load. The working principle is that a switch is connected in series with the input end of the electronic load, and when the temperature or the current of the electronic load is detected to exceed a set value, the switch is turned off, and the power supply is cut off, so that the electronic load is protected. The advantage of this circuit is that it is simple, reliable and responds quickly, but it has drawbacks such as increasing the output impedance of the power supply, affecting the stability and efficiency of the power supply, and switching noise and electromagnetic interference. Moreover, this approach has some limitations, such as the fixed threshold may not be applicable to all load situations, and problems of misjudgment or missed judgment are likely to occur. In addition, the method cannot adapt to the dynamic change of the working state of the electronic load.
The setting of the fixed threshold may lead to erroneous judgment, i.e. erroneous judgment of the normal operating state as an overheat or overcurrent state. For example, when an electronic load produces a brief peak current in the normal operating range, the switch protection circuit may erroneously determine that it is in an overcurrent condition and trigger a protection action. The setting of a fixed threshold may also lead to missed decisions, i.e. failure to correctly detect a true overheat or overcurrent condition. For example, if the electronic load is required to withstand a higher current or temperature under certain operating conditions, and the threshold setting is low, the switch protection circuit may not trigger the protection action in time, resulting in damage to the electronic load. The switch protection circuit typically sets the threshold based on statistical methods, but these outliers may lead to erroneous decisions in the face of sudden current peaks or other anomalies. For example, an electronic load may generate high current peaks at the moment of start-up or shut-down, which peaks may exceed a set threshold, but do not represent a true overcurrent condition.
To address these issues, deep learning based artificial intelligence techniques may provide a more accurate method of judgment. By analyzing the real-time current waveform of the electronic load, the deep learning algorithm can learn and identify various modes of fluctuation, including normal operating conditions and abnormal conditions. Compared with the setting of a fixed threshold, the deep learning algorithm can carry out self-adaptive adjustment according to actual conditions, and the judgment accuracy is improved.
Therefore, in order to further improve the accuracy and reliability of the switch protection circuit, the application provides an artificial intelligence technology based on deep learning to optimize the judging process, the real-time current waveform of the electronic load is analyzed, the fluctuation mode characteristic is extracted, the mode identification is carried out by using a deep learning algorithm, the more accurate abnormal condition judgment is realized, the misoperation and damage risk are avoided, and the technology can provide higher accuracy, adaptability, instantaneity and anti-interference capability, so that the performance of the switch protection circuit is further improved.
In one embodiment of the present application, fig. 1 is a block diagram of a high voltage dc electronic load switch protection circuit according to an embodiment of the present application. As shown in fig. 1, a high voltage dc electronic load switch protection circuit 100 according to an embodiment of the present application includes: a current value obtaining module 110, configured to obtain current values of the electronic load at a plurality of predetermined time points within a predetermined period of time; a vector arrangement module 120, configured to arrange the current values of the electronic loads at the plurality of predetermined time points into time-dimension time-series input vectors of the current values of the electronic loads; the image conversion module 130 is configured to perform image conversion on the electronic load current value time sequence input vector to obtain a sequence of electronic load current value local time sequence images; a time sequence feature extraction module 140, configured to extract a current fluctuation time sequence feature in the sequence of the electronic load current value local time sequence image; the switch control module 150 is configured to determine whether to turn off the switch based on the current ripple timing characteristic.
In the current value acquisition module 110, ensuring accurate acquisition of current values of the electronic load at a plurality of predetermined points in time over a predetermined period of time, it may be necessary to use appropriate sensors or measurement devices to acquire the current values and ensure that the sampling frequency is high enough to capture rapid changes in current. Real-time current data of the electronic load is provided, and input is provided for processing of subsequent modules.
In the vector arrangement module 120, the current values of the electronic loads at a plurality of predetermined time points are arranged into the time sequence input vector of the current values of the electronic loads according to the time dimension, so that the dimension and the sequence of the vector are correct, and the subsequent modules can process correctly. The current values are arranged into vectors according to the time dimension, so that time sequence information of the current values of the electronic load is provided, and a foundation is provided for processing of subsequent modules.
In the image conversion module 130, the electronic load current value time sequence input vector is subjected to image conversion, and a sequence of electronic load current value local time sequence images is generated. The appropriate image conversion method is chosen, for example to convert the vector into a grey-scale or color map, and to ensure that the conversion process does not introduce excessive information loss. The electronic load current values are converted into image sequences so that subsequent modules can analyze the time sequence characteristics of the current fluctuations using computer vision and image processing methods.
In the timing feature extraction module 140, a timing feature of the current ripple is extracted from a sequence of partial timing images of the electronic load current value. It may be desirable to use a suitable feature extraction algorithm, such as a deep learning based convolutional neural network or conventional signal processing methods, to capture the pattern and trend of current fluctuations. By extracting the time sequence characteristics of the current fluctuation, the working state of the electronic load, including the overheat or overcurrent state, can be judged more accurately.
In the switch control module 150, it is determined whether to turn off the switch based on the timing characteristics of the current ripple. And judging whether the electronic load is in an abnormal state according to a specific judging rule or a specific threshold value, and triggering a control signal for closing the switch if the electronic load is in the abnormal state. By judging based on the time sequence characteristics of current fluctuation, whether the switch needs to be turned off or not can be accurately determined, so that the problem of misjudgment or missed judgment is avoided, and the protection effect on the electronic load is improved.
The current value acquisition module, the vector arrangement module, the image conversion module, the time sequence feature extraction module and the switch control module cooperate together, and the accuracy and the reliability of the switch protection circuit are improved by utilizing the technologies of deep learning, image processing and the like, so that the electronic load is better protected from being damaged by overheat or overload.
Aiming at the technical problems, the technical concept of the application is that whether the threshold value is exceeded is not judged according to the set value simply, the real-time current waveform of the electronic load is analyzed by combining the artificial intelligence technology based on deep learning, the fluctuation mode characteristics of the electronic load are extracted, so that abnormal conditions are identified, and whether a switch is closed is judged, so that misoperation caused by outliers is avoided.
Based on this, in the technical solution of the present application, first, current values of electronic loads at a plurality of predetermined time points within a predetermined period of time are acquired; and arranging the current values of the electronic loads at a plurality of preset time points into time sequence input vectors of the current values of the electronic loads according to the time dimension.
By arranging the current values according to the time dimension, the time sequence information of the electronic load can be reserved, and the dynamic change and fluctuation mode of the current can be captured, so that the working state and the performance of the electronic load can be more comprehensively known. After the current values are arranged into the time sequence input vectors, various time sequence feature extraction methods and techniques can be more conveniently applied. For example, the time-series input vector may be extracted by using a sliding window, fourier transform, wavelet transform, or other method to extract information such as frequency, amplitude, and time domain characteristics of the current ripple. After the current values are arranged into the time series input vector, the data can be more easily processed and analyzed. For example, machine learning algorithms may be utilized to model and predict time-lapse input vectors, identify anomalies in electronic loads, or predict future current trends. In addition, statistical methods, timing analysis and other techniques can be applied to perform deeper data analysis on the timing input vector. After the current values are arranged into the time sequence input vectors, various time sequence models and algorithms can be applied for analysis and prediction. For example, a Recurrent Neural Network (RNN), long and short term memory network (LSTM), convolutional Neural Network (CNN), etc. model may be used to model and predict the time series input vector. This has the advantage that the advantage of a time-series model can be exploited to more accurately analyze and predict the current behaviour of the electronic load.
The arrangement of the current values of the electronic load at a plurality of preset time points in a preset time period into the time sequence input vector of the current value of the electronic load according to the time dimension is beneficial to retaining time sequence information, facilitating feature extraction, simplifying data processing and analysis, is widely applicable to various time sequence models and algorithms, improves understanding and analysis capability of the current behaviors of the electronic load, and provides support for optimization and reliability of a switch protection circuit.
And then, carrying out image conversion on the time sequence input vector of the electronic load current value to obtain a sequence of the partial time sequence image of the electronic load current value. Here, the time series data is converted into an image, and the spatial characteristics in the image data, the characteristics such as a current fluctuation mode and the like, and the change trend of the local waveform can be captured by utilizing a method and a technology in the field of computer vision.
In a specific example of the present application, the image conversion module is configured to perform an encoding process of converting the electronic load current value time sequence input vector into a sequence of electronic load current value local time sequence images, where the encoding process includes: the vector segmentation unit is used for carrying out vector segmentation on the electronic load current value time sequence input vector so as to obtain a sequence of the electronic load current value local time sequence input vector; and a vector-image conversion unit for inputting the sequence of the electronic load current value local time sequence input vectors through a vector-image converter respectively to obtain the sequence of the electronic load current value local time sequence images.
Wherein the vector-image conversion unit includes: the local vector segmentation subunit is used for carrying out vector segmentation on the sequence of the electronic load current value local time sequence input vector so as to obtain the sequence of the electronic load current value local input sub-vector; a matrixing subunit, configured to arrange the sequence of the local input sub-vectors of electronic load current values into a local time-sequence input matrix of electronic load current values; the normalization subunit is used for performing normalization processing on the electronic load current value local time sequence input matrix to obtain a sequence of the electronic load current value local time sequence image; wherein the range of values of each position in the sequence of the electronic load current value local time sequence image is 0-255.
The vector segmentation unit is used for segmenting the time sequence input vector of the electronic load current value and dividing the time sequence input vector into a sequence of the local time sequence input vector of the electronic load current value. This has the advantage that the whole time sequence process can be broken down into a plurality of local time sequences, thereby better capturing the local characteristics of the current ripple. By means of segmentation, the analysis precision and accuracy of current fluctuation can be improved.
The vector-image conversion unit converts the sequence of the electronic load current value local time sequence input vectors into a sequence of the electronic load current value local time sequence images through a vector-image converter respectively. The benefit of this conversion is that computer vision and image processing techniques can be used to further analyze the timing characteristics of the current fluctuations, the images provide more visual and rich information, and the patterns and trends of the current fluctuations can be better captured.
Through the combined use of the vector segmentation unit and the vector-image conversion unit, the time sequence input vector of the electronic load current value can be converted into a sequence of the local time sequence image of the electronic load current value, and the conversion process can provide more comprehensive and detailed information, so that the subsequent time sequence feature extraction module can analyze the features of the current fluctuation more accurately. Meanwhile, the image sequence can be further analyzed and identified through an image processing and computer vision method, so that the accuracy of judging the working state of the electronic load is improved.
The vector segmentation unit and the vector-image conversion unit have the beneficial effect of converting the time sequence input vector of the electronic load current value into a sequence of the local time sequence images of the electronic load current value. The conversion process can provide more comprehensive and detailed information, and more accurate input is provided for the subsequent time sequence characteristic extraction and judgment module, so that the performance and reliability of the switch protection circuit are improved.
Then, current fluctuation timing characteristics in the sequence of the electronic load current value local timing images are extracted. In particular, the electronic load current value may vary periodically. For example, for some loads, the current value may be periodically ramped up and down at a fixed frequency. When the electronic load current value abnormally fluctuates, it is often indicated that the operating state is abnormal.
In a specific example of the application, the implementation manner of extracting the current fluctuation time sequence characteristic in the sequence of the electronic load current value local time sequence image is to make the sequence of the electronic load current value local time sequence image pass through a current mode time sequence characteristic extractor based on a three-dimensional convolutional neural network model to obtain a current fluctuation time sequence characteristic diagram.
In a specific embodiment of the present application, the timing feature extraction module includes: the time sequence feature extraction unit is used for enabling the sequence of the electronic load current value local time sequence image to pass through a current mode time sequence feature extractor based on a three-dimensional convolutional neural network model so as to obtain a current fluctuation time sequence feature diagram; and a current fluctuation timing characteristic generation unit configured to take the current fluctuation timing characteristic map as the current fluctuation timing characteristic.
Wherein, the time sequence feature extraction unit is used for: and respectively carrying out convolution processing, pooling processing and nonlinear activation processing based on the three-dimensional convolution kernel on the sequence of the electronic load current value local time sequence image in forward transfer of layers by using each layer of the three-dimensional convolution neural network model-based current mode time sequence feature extractor so as to output the current fluctuation time sequence feature diagram by the last layer of the three-dimensional convolution neural network model-based current mode time sequence feature extractor.
The time sequence feature extraction unit processes the sequence of the electronic load current value local time sequence images by using a current mode time sequence feature extractor based on a three-dimensional convolutional neural network model, and extracts a time sequence feature graph of current fluctuation. By using a convolutional neural network model, patterns and trends of current fluctuations can be automatically learned and captured, thereby more accurately characterizing the current fluctuations.
The current fluctuation time series characteristic generation unit takes the current fluctuation time series characteristic diagram obtained by the time series characteristic extraction unit as a current fluctuation time series characteristic. In this way, the timing characteristics of the current ripple are translated into a more compact and advanced representation so that subsequent decision and making modules can be more conveniently used and handled.
The sequence of the partial timing images of the electronic load current values can be converted into the timing characteristics of the current fluctuations by the combined use of the timing characteristic extraction unit and the current fluctuation timing characteristic generation unit. These timing characteristics, which more accurately characterize the pattern and trend of current fluctuations, provide more rich and useful information that can be used in subsequent decision and making modules, such as determining whether an electronic load is in an abnormal state and triggering a switch control signal.
The time sequence feature extraction unit and the current fluctuation time sequence feature generation unit have the advantages that the time sequence features of the current fluctuation are extracted and generated, the time sequence features can more accurately represent the mode and trend of the current fluctuation, and more abundant and useful information is provided for subsequent judgment and decision modules, so that the performance and reliability of the switch protection circuit are improved.
And then, the current fluctuation time sequence characteristic diagram passes through a channel attention module to obtain a channel visualization current fluctuation time sequence characteristic diagram. Here, in the current fluctuation timing characteristic diagram, each channel corresponds to different timing characteristic information. However, not all feature channels are of equal importance for classification tasks. Some channels may contain noise or redundant information, while some channels may contain more distinguishing key features. By introducing a channel attention module, the weight of each channel can be automatically learned, focusing more attention on the characteristic channels that are more useful for classification tasks. Thus, the expression capability of useful features can be enhanced, and the dependence on useless features is reduced, so that the accuracy and the robustness of classification are improved. And then, the optimized channel visualization current fluctuation time sequence characteristic diagram is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating whether a switch is closed or not.
In a specific embodiment of the present application, the switch control module includes: the channel attention unit is used for passing the current fluctuation time sequence characteristic diagram through the channel attention module to obtain a channel visualization current fluctuation time sequence characteristic diagram; the characteristic distribution optimizing unit is used for carrying out characteristic distribution optimization on the channel visualization current fluctuation time sequence characteristic diagram so as to obtain an optimized channel visualization current fluctuation time sequence characteristic diagram; and the classification unit is used for enabling the optimized channel visualization current fluctuation time sequence characteristic diagram to pass through a classifier to obtain a classification result, and the classification result is used for indicating whether a switch is closed or not.
Wherein, the passageway attention unit is used for: inputting the current fluctuation time sequence characteristic diagram into a plurality of convolution layers of the channel attention module to obtain a current convolution characteristic diagram; calculating the global average value of each feature matrix of the current convolution feature diagram along the channel dimension to obtain a channel feature vector; inputting the channel feature vector into the Sigmoid activation function to obtain a channel attention weight vector; and respectively weighting each characteristic matrix of the current convolution characteristic diagram along the channel dimension by taking the characteristic value of each position in the channel attention weight vector as a weight to obtain the channel visualization current fluctuation time sequence characteristic diagram.
The channel attention unit processes the current fluctuation time sequence characteristic diagram through the channel attention module so as to obtain a channel visualization current fluctuation time sequence characteristic diagram. The channel attention module is capable of automatically learning and adjusting the weight of each channel so that proper attention and emphasis can be given to different characteristic channels of the current fluctuation, and thus, the identification and utilization of key characteristics can be improved, and the characteristics of the current fluctuation can be more accurately represented.
The characteristic distribution optimizing unit performs characteristic distribution optimization on the channel visualization current fluctuation time sequence characteristic diagram to obtain an optimized channel visualization current fluctuation time sequence characteristic diagram. By optimizing the feature distribution, the features among different channels are more balanced and effective, and the problem that some channels are too prominent or unobvious is avoided, so that the characterization capability and stability of the overall features can be improved, and further subsequent classification and decision tasks are better supported.
The classification unit processes the optimized channel visualization current fluctuation time sequence characteristic diagram through a classifier to obtain a classification result which is used for indicating whether the switch is closed or not. The classifier may be a trained machine learning model, such as a support vector machine, random forest or deep learning model, or the like. By classifying by utilizing the optimized characteristic diagram, whether the electronic load is in an abnormal state or not can be judged more accurately, so that a corresponding switch control signal is triggered.
The analysis and judgment capability of the current fluctuation time sequence characteristics can be improved through the combined use of the channel attention unit, the characteristic distribution optimizing unit and the classifying unit; the channel attention unit and the feature distribution optimizing unit can enhance the characterization and distinguishing capabilities of key features and eliminate the imbalance problem among the features; the classification unit classifies the electronic load by using the optimized feature map, and improves the accuracy and reliability of the working state of the electronic load. The channel attention unit, the characteristic distribution optimizing unit and the classifying unit have the advantages of enhancing the characteristic characterization capability, optimizing the characteristic distribution, realizing accurate judgment of the working state of the electronic load through the classifier, improving the performance and the reliability of the switch protection circuit and ensuring the safe operation of the electronic load.
Further, in a specific embodiment of the present application, the feature distribution optimizing unit is configured to: performing probability density convergence optimization of feature scale constraint on each feature matrix of the channel visualization current fluctuation time sequence feature diagram to obtain a first weight and k second weights; matrix multiplying the first weight and each second weight to obtain a sequence of weight values; and weighting each feature matrix of the channel visualization current fluctuation time sequence feature map by taking the sequence of the weight values as the weight values to obtain the optimized channel visualization current fluctuation time sequence feature map.
In the technical scheme of the application, vector segmentation is performed on the electronic load current value time sequence input vector to obtain a sequence of electronic load current value local time sequence input vector, the sequence of electronic load current value local time sequence input vector is respectively processed through a vector-image converter to obtain a sequence of electronic load current value local time sequence image, and after the sequence of electronic load current value local time sequence image is processed through a current mode time sequence feature extractor based on a three-dimensional convolutional neural network model, a local time domain based on vector segmentation of the electronic load current value in a global time domain and a time sequence correlation feature between time domains in a time domain subspace and a time domain subspace under a multiscale time domain subspace of a subdivision time domain based on a local time domain of vector-image format conversion can be extracted. Thus, the time-series correlated feature distribution in the local time-domain subspace in some local time domain can be further enhanced by the channel attention module, but if the channel-developed current fluctuation time-series feature map is taken as a whole, there may be imbalance in the time-series correlated feature expression in each time-domain subspace, and the applicant of the present application further finds that such imbalance is largely related to the feature expression scale, namely, the sub-division time-domain-inter-time-series correlated feature expression scale in the local time domain in the spatial dimension of the feature matrix, and the local time-domain-time-series correlated scale in the channel dimension between each feature matrix, for example, it can be understood that the imbalance in the distribution between the feature values with respect to the predetermined scale is more unbalanced, the overall expression of the feature map is also more unbalanced.
Thus, it is preferable that each feature matrix of the current fluctuation timing feature map is developed for the channel, for example, asPerforming probability density convergence optimization of feature scale constraint, wherein the probability density convergence optimization is expressed as: carrying out probability density convergence optimization of feature scale constraint on each feature matrix of the channel visualization current fluctuation time sequence feature diagram by using the following optimization formula; wherein, the optimization formula is:
wherein,is the characteristic matrix of each channel visualization current fluctuation time sequence characteristic diagram,/and a method for generating the same>Is the channel number of the channel visualization current fluctuation time sequence characteristic diagram, <>Is each characteristic matrix of the channel visualization current fluctuation time sequence characteristic diagramGlobal feature mean,/, of>Is->Component feature vector, < >>Representing feature vector +.>Square of the two norms of +.>Is the characteristic matrix of each channel visualization current fluctuation time sequence characteristic diagram>Is the dimension of (i.e. width multiplied by height) and +.>Each feature matrix representing the channel visualization current fluctuation time sequence feature diagram>Is the square of the Frobenius norm, < >>Is the characteristic matrix of each channel visualization current fluctuation time sequence characteristic diagram>Characteristic value of each position +.>Is a feature vector +.>Weight coefficient of>Is the characteristic matrix of each channel visualization current fluctuation time sequence characteristic diagram>Weight coefficient of (c) in the above-mentioned formula (c).
Here, the probability density convergence optimization of the feature scale constraint can perform a multi-level distribution structure correlation constraint on the feature probability density distribution in the high-dimensional feature space based on the feature scale by using a tail distribution strengthening mechanism of a quasi-standard cauchy distribution type so as to haveThe probability density distribution of the high-dimensional features with different scales is subjected to uniformity expansion in the whole probability density space, so that probability density convergence heterogeneity caused by feature scale deviation is compensated. Thus, the weight is as aboveFor each feature matrix along the channel +.>And weighting is carried out, so that the convergence of the optimized channel visualization current fluctuation time sequence characteristic diagram relative to the preset regression probability can be improved, and the accuracy of a classification result obtained by the classifier is improved.
In summary, the hvth electronic load switch protection circuit 100 according to the embodiment of the application is illustrated, which does not simply determine whether the threshold value is exceeded according to the set value, but extracts the fluctuation mode characteristics of the electronic load by analyzing the real-time current waveform of the electronic load in combination with the artificial intelligence technology based on deep learning, so as to identify the abnormal situation, thereby determining whether to turn off the switch to avoid the misoperation caused by the outlier.
In one embodiment of the present application, fig. 2 is a flowchart of a method for protecting a hvdc electronic load switch according to an embodiment of the present application. Fig. 3 is a schematic diagram of a protection method architecture for a hvdc electronic load switch according to an embodiment of the present application. As shown in fig. 2 and 3, the method for protecting the high-voltage direct-current electronic load switch includes: 210, acquiring current values of electronic loads at a plurality of preset time points in a preset time period; 220, arranging the current values of the electronic loads at a plurality of preset time points into time sequence input vectors of the current values of the electronic loads according to a time dimension; 230, performing image conversion on the electronic load current value time sequence input vector to obtain a sequence of electronic load current value local time sequence images; 240, extracting current fluctuation time sequence characteristics in the sequence of the electronic load current value local time sequence images; 250, determining whether to close the switch based on the current ripple timing characteristics.
It will be appreciated by those skilled in the art that the specific operation of the steps in the above-described hvth electronic load switch protection method has been described in detail in the above description of the hvth electronic load switch protection circuit with reference to fig. 1, and thus, repetitive description thereof will be omitted.
Fig. 4 is an application scenario diagram of a hvth electronic load switch protection circuit according to an embodiment of the application. As shown in fig. 4, in this application scenario, first, current values of an electronic load at a plurality of predetermined time points within a predetermined period of time are acquired (e.g., C as illustrated in fig. 4); the obtained current value is then input into a server (e.g., S as illustrated in fig. 4) where the hvth electronic load switch protection circuit algorithm is deployed, wherein the server is capable of processing the current value based on the hvth electronic load switch protection circuit algorithm to determine whether to turn off the switch.
The basic principles of the present application have been described above in connection with specific embodiments, however, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are merely examples and not limiting, and these advantages, benefits, effects, etc. are not to be considered as necessarily possessed by the various embodiments of the present application. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, as the application is not intended to be limited to the details disclosed herein as such.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.
Claims (8)
1. A high voltage dc electronic load switch protection circuit, comprising:
a current value acquisition module for acquiring current values of the electronic load at a plurality of preset time points in a preset time period;
the vector arrangement module is used for arranging the current values of the electronic loads at a plurality of preset time points into time sequence input vectors of the current values of the electronic loads according to the time dimension;
the image conversion module is used for carrying out image conversion on the electronic load current value time sequence input vector so as to obtain a sequence of electronic load current value local time sequence images;
the time sequence feature extraction module is used for extracting current fluctuation time sequence features in the sequence of the electronic load current value local time sequence images;
and the switch control module is used for determining whether to close the switch or not based on the current fluctuation time sequence characteristic.
2. The high voltage dc electronic load switch protection circuit of claim 1, wherein the image conversion module comprises:
the vector segmentation unit is used for carrying out vector segmentation on the electronic load current value time sequence input vector so as to obtain a sequence of the electronic load current value local time sequence input vector; and
and the vector-image conversion unit is used for respectively inputting the sequence of the electronic load current value local time sequence input vector through a vector-image converter to obtain the sequence of the electronic load current value local time sequence image.
3. The high voltage dc electronic load switch protection circuit according to claim 2, wherein the vector-image conversion unit comprises:
the local vector segmentation subunit is used for carrying out vector segmentation on the sequence of the electronic load current value local time sequence input vector so as to obtain the sequence of the electronic load current value local input sub-vector;
a matrixing subunit, configured to arrange the sequence of the local input sub-vectors of electronic load current values into a local time-sequence input matrix of electronic load current values; and
the normalization subunit is used for carrying out normalization processing on the local time sequence input matrix of the electronic load current value so as to obtain a sequence of the local time sequence image of the electronic load current value; wherein the range of values of each position in the sequence of the electronic load current value local time sequence image is 0-255.
4. The hvth electronic load switch protection circuit of claim 3, wherein the timing feature extraction module comprises:
the time sequence feature extraction unit is used for enabling the sequence of the electronic load current value local time sequence image to pass through a current mode time sequence feature extractor based on a three-dimensional convolutional neural network model so as to obtain a current fluctuation time sequence feature diagram; and
and the current fluctuation time sequence characteristic generating unit is used for taking the current fluctuation time sequence characteristic diagram as the current fluctuation time sequence characteristic.
5. The hvth electronic load switch protection circuit according to claim 4, wherein the timing feature extraction unit is configured to:
and respectively carrying out convolution processing, pooling processing and nonlinear activation processing based on the three-dimensional convolution kernel on the sequence of the electronic load current value local time sequence image in forward transfer of layers by using each layer of the three-dimensional convolution neural network model-based current mode time sequence feature extractor so as to output the current fluctuation time sequence feature diagram by the last layer of the three-dimensional convolution neural network model-based current mode time sequence feature extractor.
6. The high voltage dc electronic load switch protection circuit of claim 5, wherein the switch control module comprises:
the channel attention unit is used for passing the current fluctuation time sequence characteristic diagram through the channel attention module to obtain a channel visualization current fluctuation time sequence characteristic diagram;
the characteristic distribution optimizing unit is used for carrying out characteristic distribution optimization on the channel visualization current fluctuation time sequence characteristic diagram so as to obtain an optimized channel visualization current fluctuation time sequence characteristic diagram; and
and the classification unit is used for enabling the optimized channel visualization current fluctuation time sequence characteristic diagram to pass through a classifier to obtain a classification result, and the classification result is used for indicating whether a switch is closed or not.
7. The hvth electronic load switch protection circuit of claim 6, wherein the channel attention unit is configured to:
inputting the current fluctuation time sequence characteristic diagram into a plurality of convolution layers of the channel attention module to obtain a current convolution characteristic diagram;
calculating the global average value of each feature matrix of the current convolution feature diagram along the channel dimension to obtain a channel feature vector;
inputting the channel feature vector into the Sigmoid activation function to obtain a channel attention weight vector; and
and respectively weighting each characteristic matrix of the current convolution characteristic diagram along the channel dimension by taking the characteristic value of each position in the channel attention weight vector as a weight to obtain the channel visualization current fluctuation time sequence characteristic diagram.
8. The high voltage direct current electronic load switch protection circuit according to claim 7, wherein the characteristic distribution optimizing unit is configured to:
performing probability density convergence optimization of feature scale constraint on each feature matrix of the channel visualization current fluctuation time sequence feature diagram to obtain a first weight and k second weights;
matrix multiplying the first weight and each second weight to obtain a sequence of weight values; and
and taking the sequence of the weight values as the weight values, and weighting each feature matrix of the channel visualization current fluctuation time sequence feature map to obtain the optimized channel visualization current fluctuation time sequence feature map.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311422400.7A CN117498262B (en) | 2023-10-31 | High-voltage direct-current electronic load switch protection circuit |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311422400.7A CN117498262B (en) | 2023-10-31 | High-voltage direct-current electronic load switch protection circuit |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117498262A true CN117498262A (en) | 2024-02-02 |
| CN117498262B CN117498262B (en) | 2024-10-25 |
Family
ID=
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117805607A (en) * | 2024-02-29 | 2024-04-02 | 山西漳电科学技术研究院(有限公司) | DC level difference matching test method for power plant DC system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104794474A (en) * | 2015-03-26 | 2015-07-22 | 同济大学 | Comprehensive maximum current density vector feature extraction method based on magnetocardiogram |
| US20200284883A1 (en) * | 2019-03-08 | 2020-09-10 | Osram Gmbh | Component for a lidar sensor system, lidar sensor system, lidar sensor device, method for a lidar sensor system and method for a lidar sensor device |
| CN114475350A (en) * | 2022-02-14 | 2022-05-13 | 杭州鸽然科技有限公司 | Intelligent charging system and working method thereof |
| CN116299051A (en) * | 2023-05-19 | 2023-06-23 | 深圳市安科讯电子制造有限公司 | Overcurrent short circuit detection circuit and detection protection system |
| CN116524200A (en) * | 2023-05-05 | 2023-08-01 | 东南大学 | High-voltage circuit breaker fault diagnosis method based on image recognition |
| CN116683648A (en) * | 2023-06-13 | 2023-09-01 | 浙江华耀电气科技有限公司 | Intelligent power distribution cabinet and control system thereof |
| CN116934304A (en) * | 2023-08-03 | 2023-10-24 | 广东省俊耀控股有限公司 | Intelligent power distribution room equipment operation maintenance management system and method thereof |
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104794474A (en) * | 2015-03-26 | 2015-07-22 | 同济大学 | Comprehensive maximum current density vector feature extraction method based on magnetocardiogram |
| US20200284883A1 (en) * | 2019-03-08 | 2020-09-10 | Osram Gmbh | Component for a lidar sensor system, lidar sensor system, lidar sensor device, method for a lidar sensor system and method for a lidar sensor device |
| CN114475350A (en) * | 2022-02-14 | 2022-05-13 | 杭州鸽然科技有限公司 | Intelligent charging system and working method thereof |
| CN116524200A (en) * | 2023-05-05 | 2023-08-01 | 东南大学 | High-voltage circuit breaker fault diagnosis method based on image recognition |
| CN116299051A (en) * | 2023-05-19 | 2023-06-23 | 深圳市安科讯电子制造有限公司 | Overcurrent short circuit detection circuit and detection protection system |
| CN116683648A (en) * | 2023-06-13 | 2023-09-01 | 浙江华耀电气科技有限公司 | Intelligent power distribution cabinet and control system thereof |
| CN116934304A (en) * | 2023-08-03 | 2023-10-24 | 广东省俊耀控股有限公司 | Intelligent power distribution room equipment operation maintenance management system and method thereof |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117805607A (en) * | 2024-02-29 | 2024-04-02 | 山西漳电科学技术研究院(有限公司) | DC level difference matching test method for power plant DC system |
| CN117805607B (en) * | 2024-02-29 | 2024-05-07 | 山西漳电科学技术研究院(有限公司) | DC level difference matching test method for power plant DC system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yi et al. | Line-to-line fault detection for photovoltaic arrays based on multiresolution signal decomposition and two-stage support vector machine | |
| Harrou et al. | Improved $ k $ NN-based monitoring schemes for detecting faults in PV systems | |
| Chen et al. | Data-driven detection of hot spots in photovoltaic energy systems | |
| Appiah et al. | Long short-term memory networks based automatic feature extraction for photovoltaic array fault diagnosis | |
| Li et al. | Fast power system event identification using enhanced LSTM network with renewable energy integration | |
| Ahmadi et al. | A new method for detecting series arc fault in photovoltaic systems based on the blind-source separation | |
| CN111444615B (en) | Photovoltaic array fault diagnosis method based on K nearest neighbor and IV curve | |
| CN114825257B (en) | Leakage protection device for LED lamp detection and leakage protection method thereof | |
| Zhu et al. | Networked time series shapelet learning for power system transient stability assessment | |
| CN117110798B (en) | Fault detection method and system for intelligent power distribution network | |
| Sekar et al. | An improved power quality disturbance detection using deep learning approach | |
| CN117612345A (en) | Power equipment state monitoring and alarming system and method | |
| CN109039280B (en) | Photovoltaic array fault diagnosis method based on non-principal component data characteristics | |
| Zhu et al. | Integrated data-driven power system transient stability monitoring and enhancement | |
| CN117269644A (en) | Line fault monitoring system and method for current transformer | |
| Fan et al. | Post-fault transient stability assessment based on k-nearest neighbor algorithm with Mahalanobis distance | |
| CN117498262B (en) | High-voltage direct-current electronic load switch protection circuit | |
| CN117933549A (en) | Enterprise power consumption data processing method and system based on power threshold calculation algorithm | |
| CN117621875A (en) | Electric automobile charging management system and method based on low-temperature compensation | |
| Afrasiabi et al. | Machine Learning-Based Condition Monitoring of Solar Photovoltaic Systems: A Review | |
| CN117498262A (en) | High-voltage direct-current electronic load switch protection circuit | |
| Ahmad et al. | Prediction of cascading failures and simultaneous learning of functional connectivity in power system | |
| Del Río et al. | Photovoltaic Solar Power Plant Maintenance Management based on IoT and Machine Learning | |
| Trinh et al. | Application of Artificial Intelligence in Detecting and Classifying Faults of Solar Panels | |
| Liu et al. | Fault diagnosis of photovoltaic array based on XGBoost method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |